IBM Copy Services Manager (CSM) should be differentiate between real HOST-Connections issues vs. a planned shut down from one or more HOST systems.

Starting with CSM 6.3.10 IBM has introduced his Multi Sysplex Support (feature 87578b). With this feature are several resilience topics addressed to make the tool less prone to incorrect operation. One decision that has been made, for example, is that the HyperSwap Command isn’t available at the GUI when one HOST-Connection is in disconnected status. The design is, that a planned HyperSwap shouldn’t be supplied to the customer to prevent him to do any failures during a loss of one HOST-Connection. A second observation with an interrupted HOST-Connection was that all MM FO/FB Sessions which are enabled for Hardened Freeze Operations go to severe status (IWNR7008E).

For a real communication issue scenario like HOST-connection is dropped but the z/OS LPAR still runs this enhancement is good solution to prevent customers for incorrect operations. However, more and more regulatory requirements come up that a customer should provide a proof that he can handle its workload if one (ore more) z/OS LPARs aren’t available. Such evidence must be provided over several days or even weeks are possible. For this timeframe it is a “normal situation” that a HOST-Connection is in disconnected status.

The IDEA should address this requirement that CSM is able to distinguish between an Outage of the HOST-Connection vs. a planned shut down from one LPAR. If one LPAR left the Sysplex CSM should ignore this/these HOST-Connection(s) and should enable the HyperSwap Command for HS managed Sessions. And as well the regain HSIB Connection behavior should also be enhanced, that MM FO/FB Sessions also go back to “normal status” when the connection lost to one HOST is a planned situation like described above.